This in-frequently modeled rocket remains a favorite of many modelers who build and fly Saturn Launch Vehicles.

Like most major development projects, the evolution of the Saturn I changed between conception and execution, although the configuration that emerged in 1958 was subjected to remarkably few major design variations before its first launch in 1961. The basic outlines for ABMA's concepts of the Saturn I (when it was still called Juno V) were sketched out in two reports to Advanced Research Projects Agency (ARPA) in October and November 1958; insights on various aspects of early design choices were provided by von Braun himself in ABMA's presentation to NASA in December 1958. For example, original concepts for yaw, pitch, and roll control called for hinged outer engines: two hinged for pitch; two hinged for yaw; all four for roll. But application of adequate control forces required fairly high deflection of the engine thrust vector, and the engine contractor (Rocketdyne) complained that this would put too much stress on propellant flex lines. Instead, gimbaling of all four outer engines was adopted, achieving adequate control force with less engine deflection. The gimbal system for mounting engines permitted each engine in the cluster to swivel about for either yaw or pitch control.

On the other hand, the original multiengine concept was maintained. Throughout the early design phase, ABMA stressed the reliability of the multiengine approach in case one or even two were lost. Particularly in the case of manned missions, von Braun emphasized, the engine-out capability offered much higher margins of safety in continuing a mission until conditions were less hazardous for separation of the crew capsule.

The multitank design also persisted as a design choice. In his NASA presentation, von Braun praised the multitank design for several reasons. Component tanks could be flown by Douglas C-124 Globemasters to any part of the world and reassembled for launch; this procedure would provide a high degree of flexibility. The separate tanks eliminated the technical difficulties of internal horizontal bulkheads, required in a large tank vehicle, to keep fuel and oxidizer separate. It also meant a shorter, and more desirable, vehicle. In spite of the added weight, most rocket propellant tanks included internal fuel slosh baffles, because splashing and surging of the liquid fuel created problems in keeping the vehicle stable and under control. In 1958, von Braun predicted that no fuel slosh baffles would be required in the multitank design because of the small diameter of the individual tanks (although the flight versions actually incorporated slosh baffles in their design). A great deal of attention was also given to booster recovery schemes, in which the spent first stage would be recovered from the ocean after its descent had been slowed by retrorockets and parachutes. The Huntsville group foresaw immense savings in the recovery scheme, since the illustration given by von Braun assumed "5 or 19 years from now" a launch rate of 100 vehicles per year over a 5-year period, at a cost of about $10 million per launch.39

[77] More than any of the Saturn vehicles, the Saturn I S-I stage configuration evolved during flight tests (for details, see chapter 11). NASA developed the Saturn I as first-generation and second-generation rockets, designated Block I and Block II. The first four launches used the Block I vehicle, with inert upper stages and no fins on the first stage, the S-I. Block II versions carried a live second stage, the S-IV, sported a corolla of aerodynamic fins at the base, and used uprated H-I engines. The S-I first stage for the Saturn I also became the first stage of the Saturn IB; in this application, it was called the S-IB. Again, there were modifications to the fins, engines, and various internal components. Nevertheless, the basic details of fabrication and testing of the Saturn I and Saturn IB remained similar. The first stage of the Saturn I and IB may have looked like a plumber's nightmare, but it fit the criteria of conservative design and economy established early in the program. As Marshall engineers discovered, development of a new booster of Saturn I's size involved a number of design problems. Fabrication of the tankage was comparatively easy. Even though the former Redstone and Jupiter tanks had to be lengthened from 12 to 16 meters to carry added propellants, the basic diameters of the 178-centimeter Redstone and 267-centimeter Jupiter tanks were retained, so they could be fabricated from the tooling and welding equipment still available at Huntsville. The tank arrangement settled on by MSFC gave an alternate pattern of the four fuel and four oxidizer tanks, clustered around the 267-centimeter center oxidizer tank. The oxidizer tanks carried the load from the upper stages of the Saturn, the fuel tanks only contributing to the lateral stiffness of the cluster. When filled, the oxidizer tanks contracted 63.5 millimeters, which meant that the fuel tanks had to have slip joints at their upper ends to accommodate other structural elements that fluctuated with the tank shrinkage. All together, the Saturn I first stage carried 340 000 kilograms of propellants in its nine tanks. To keep the propellant in one tank from depleting too rapidly during flight, which would seriously unbalance the vehicle, the Saturn I incorporated an interconnecting pipe system, with regulating equipment to keep propellants at uniform level in all tanks during a mission. Each of the four outboard fuel tanks fed two engines, yet interconnected with the other tanks. The 267-centimeter center liquid-oxygen (LOX) tank provided series flow to the four outboard LOX tanks, which also fed two engines apiece.

Although the group of tanks eased the potential slosh tendencies of a single large tank, each separate cylinder contained fixed baffles, running accordionlike down the tank interiors. Pressurization for the LOX tanks was done by a heat exchanger, dumping it into the top of the LOX tanks as gaseous oxygen. Gaseous nitrogen from fiberglass spheres at the top of the booster pressurized the fuel tanks. The 48 spheres fixed to the top of the stage were curiously reminiscent of bunches of grapes.

The cluster of tanks was held together at the base by the tail section [78] and at the top by an aptly named structural component known as the "spider beam." The tail section consisted of the thrust structure assembly as well as the heat shield, shrouding for engine components, holddown points, stabilizing fins (on the later Saturn I first stages), and other components.

Assembly of the spider beam required a special fixture for precise alignment and joining of the heavy aluminum l beams, of which it was made. Starting with a hub assembly, eight radial beams were attached to it at 45-degree intervals. Then eight more cross beams were joined to the outer ends of the radials with splice plates. The spider beam played a dual role. Special hardware attached to it was used during the initial clustering of the tanks. In other words, the spider beam served as an assembly fixture, then remained as part of the stage's permanent structural assemblies, with each outboard oxidizer tank affixed to the beam. Because a smaller diameter upper stage of 5.6 meters was planned for the Saturn I, an upper shroud was incorporated as part of the structural transition from the larger 6.5-meter-diameter first stage. The upper shroud also enclosed telemetry equipment, umbilical connection points used in ground test and launch preparation, and space for the recovery system for the first stage. In the later versions (the Block II models), the shroud section was eliminated, and instruments were housed in a separate instrument segment atop the upper stage. The recovery section was no longer required; additional studies, completed by early 1962, indicated that the recovery scheme would require extensive modification to the stage, so the idea was finally dropped.40

In the process of refining the design of the Saturn I, two major problems emerged: stability and base heating. As with most large rockets, the Saturn I was highly unstable, with the overall center of gravity located on the heavy, lower-stage booster, while the center of lift, in most flight conditions, was high on the upper stages. The nature of the problem called for more advanced control processes than used on aircraft and rockets the size of ICBMs. The low natural frequency of the big vehicle was such that when the gimbaled engines moved to correct rocket motions, special care had to be taken not to amplify the motions because the control system frequency was close to that of the vehicle itself.

More worrisome, at least in the early design stage, was the problem of base heating. Even with a rocket powered by only one engine, the flow pattern at its base proved nearly impossible to predict for the various combinations of speed and altitude. Base heating occurred when the rocket exhaust interacted with the shock waves trailing behind the vehicle. This clash created unpredictable regions of dead air and zones of turbulent mixing. Heated by the rocket exhaust, the air trapped in these areas in turn raised the heat levels at the base of the rocket to undesirable temperatures. Worse, the fuel-rich exhaust flow from the engine turbopump could get caught in these "hot-spot" regions, causing fire or explosion.

[79] The base heating phenomenon became worse with multiengine rockets. The eight-engine Saturn I cluster began to look like a Pandora's box of base heating. To get an idea of what to expect, and to work out some fixes ahead of time, the Saturn design team ran some cold flow tests, using scale-model hardware, and called on NASA's Lewis Research Center, in Cleveland, to run some unusual wind tunnel tests. These investigations involved a booster model with eight operating engines, each putting out 1100 newtons (250 pounds) of thrust. Following the tests and extensive theoretical studies, designers in Huntsville came up with several ideas to cope with the base-heating situation. Arranged in a cross-shaped configuration, the engine pattern of the cluster was conceived to minimize dead air regions and turbulent zones. The four inner engines were bunched together in the center to reduce excessive heating in the central area, and the remaining four were positioned to avoid structural interference as the gimbaled engines swung on their mounts. The lower skirt was designed to direct large streams of high-energy air toward the four center engines in particular to prevent dead air regions from developing in their vicinity. A heavy fire wall was installed across the base of the booster near the throat of the engines, with flexible engine skirts to permit gimbaling and, at the same time, keep the super-heated gas from flowing back up to the turbopumps and propellant lines above. The problem of the exhaust from the turbopumps received special attention. For the four center engines, which were fixed, the fuel-rich exhaust gases were piped to the edge of the booster skirt and dumped overboard into a region of high-velocity air flow. In later vehicles, the exhaust gases were dumped exactly into the "centerstar" created by the four fixed engines. The gimbaled outboard engines required a different approach. The turbopump was fixed to the gimbaled engines; therefore an overboard duct for them would have required a flexible coupling that could withstand the high temperatures of the turbine exhaust gases. Instead, MSFC devised outboard engine attachments called aspirators, which forced the turbine exhaust into hoods around the engine exhaust area and mixed the turbopump exhaust with the engine's main exhaust flow.41

Successful ignition and operation of an eight-engine cluster of Saturn's dimensions required extensive testing beforehand. In December 1958, ARPA released funds for modifications to one side of a two-position Juno test tower in order to test-fire the Saturn I first stage. Preparations for these static tests, as they were called, required extensive reworking of the Saturn's side of the tower, including a new steel and concrete foundation down to bedrock, a steel overhead support structure and a 110-metric ton overhead crane, a new flame deflector and fire-control system, and much new instrumentation. The job took a whole year. By January 1959, ABMA crews installed a full-sized, high-fidelity [80] mockup of the first stage in the tower to check all the interfaces for service and test equipment. Satisfied, they took the mockup out, and put in the first static-test version. The test booster, SA-T, was installed during February, and late in March the first firing test, a timid one, burned only two engines for an eight-second run. Many skeptics still doubted that the eight-engine cluster would operate satisfactorily. "People at that time still had a lot of difficulty persuading individual rocket motors to fire up... reliably," von Braun explained, "and here we said we would fire up all eight simultaneously." There were a lot of jokes about "Cluster's Last Stand," von Braun chuckled. Still, the firing crew at Marshall proceeded cautiously. Not until the third run, on 29 April 1960, did test engineers fire up all eight barrels, and then only for an eight-second burst. By the middle of June, the first stage was roaring at full power for more than two minutes.

Reverberations of the Saturn tests were quickly felt. The acoustical impact was quite evident in the immediate area around the city of Huntsville, and the long-range sound propagation occurred at distances up to 160 kilometers. The result was a rash of accidental damage to windows and wall plaster, followed by a rash of damage claims (sometimes filed by citizens on days when no tests had been conducted). Aware that climatic conditions caused very pronounced differences in noise levels and long-range sound propagation, engineers began taking meteorological soundings and installed a huge acoustical horn atop a tower in the vicinity of the test area. No ordinary tooter, the horn was over 7.6 meters long and had a huge flared aperture over 4.6 meters high. Its sonorous gawps, bounced off a network of sound recorders, gave acoustical engineers a good idea whether it was safe to fire the big rockets on overcast days.42

To make the most use of the expensive test facilities, as soon as a booster completed its test-firing series and was shipped off to Cape Canaveral for launch, the SA-T booster was fastened back into place for further verification and testing of Saturn systems. The complex test instrumentation was complemented by the growing sophistication of automatic checkout systems used in the Saturn I first stage. Early hardware was designed for manual checkout. As more advanced electronics and computers became available, significant portions of the procedure were designed for automatic tests and checks. The scope of automatic test and checkout evolved into a complex network that tied together diverse geographic test and manufacturing locations. Later generations of Saturn vehicles and individual components were electronically monitored, literally, from the time of the first buildup on the shop floor until the mission was finished in outer space.

Because manufacturing tests of individual stages occurred separately at diverse locations, a specialized facility was required to verify the physical interface design, system integration, and system operation of the total vehicle. During a flight, natural structural frequencies occurred-the result of vibrations of moving parts, aerodynamic forces, and soon. If the control-force input of gimbaling engines, for example, reinforced the structure's natural frequency, the amplification of such structural deflections could destroy the vehicle. So a dynamic test stand, large enough to surround a complete two-stage Saturn I, was begun at MSFC in the summer of 1960 and finished early in 1961. The dynamic test facility was designed to test the vehicle either in entirety or in separate flight configurations. Vibration loads could be applied to the vehicle in pitch, yaw, roll, or longitudinal axis to get data on resonance frequencies and bending modes. Saturn I tests uncovered several problem areas that were then solved before launch. Matching frequencies in the gimbal structure and hydraulic system were uncovered and "decoupled." Static tests revealed weaknesses in the heat-shield curtains around the engines, so the flexible curtains were redesigned. Structural failure of the outer liquid-oxygen tanks required a reworking of the propellant flow system.43

Historically, the style of ABMA operations emphasized in-house fabrication and production, as Army arsenals had traditionally done. As the scale of the Saturn program increased, MSFC made the obvious and logical choice to turn over fabrication and manufacture to private industry. At the same time, the center retained an unusually strong in-house capability, to keep abreast of the state of the art, undertake preliminary work on new prototype hardware, and to make sure that the contractor did the job properly.  The do-it-yourself idea was most strongly reflected in the development of the Saturn I first stage. Ten Saturn I vehicles were built and launched; the first eight used S-I first stages manufactured by MSFC, although the fifth flight vehicle carried a contractor-built second stage (the Douglas S-IV).  The last two Saturn Is to be launched had both stages supplied by private industry. Douglas supplied the S-IV upper stage, and the Chrysler Corporation's Space Division supplied the S-I lower stage.

Late in the summer of 1961, while the first Saturn I was en route to Florida for launch, MSFC began plans to select the private contractor to take over its S-I stage. The manufacturing site at Michoud was announced on 7 September, and a preliminary conference for prospective bidders occurred in New Orleans on 26 September. The first Saturn I was launched successfully one month later (27 October), and on 17 November, Chrysler was selected from five candidates to produce the S-I first stage. The final contract called for the manufacture, checkout, and test of 20 first-stage boosters. Chrysler participated in the renovation of Michoud as it tooled up for production. In the meantime, the shops at Marshall turned out the last seven S-I boosters, progressively relinquishing the primary production responsibility.

Source:  Bilstein, Roger E. (1980). Stages to Saturn: A Technological History of the Apollo/Saturn Launch Vehicles. NASA SP-4206. ISBN 0-16-048909-1.